Parkinson’s Disease (PD) is considered one of the most complex and challenging neurodegenerative disorders of our time. It affects nearly one million people in the United States alone, with approximately 90,000 new diagnoses each year. Characterized by progressive motor symptoms such as tremor, rigidity, and bradykinesia, as well as a host of non-motor symptoms, Parkinson’s presents with a wide range of manifestations and variable progression patterns. This complexity makes it particularly difficult to develop universally effective treatments.
Despite decades of research and clinical advancements, we have yet to fully understand the disease’s underlying mechanisms, and we continue to grapple with the limitations of traditional research models. Clinical trials remain the gold standard for evaluating new treatments, yet they face significant hurdles—high costs, time-consuming recruitment processes, and often, a lack of generalizability due to narrow inclusion criteria. Many trials exclude older adults or those with coexisting medical conditions, resulting in study populations that don’t accurately reflect the broader PD community.
Leveraging the vast potential of real-world data (RWD) and artificial intelligence (AI) has the potential to overcome these challenges and truly transform PD research to bring earlier diagnoses, more personalized treatments, and more efficient therapeutic development.
Real-World Data: A Transformative Asset for Parkinson’s Research
Traditional clinical trials offer a valuable, but limited, snapshot of the patient experience. In contrast, RWD—collected from sources such as electronic health records (EHRs) of specialty clinical registries—provide a more comprehensive, longitudinal view of a patient’s health journey. Specialty clinical registries, in particular, offer rich, disease-specific datasets that help illuminate patterns that might be missed in controlled trial settings.
By analyzing real-world evidence (RWE), derived from RWD, life sciences companies gain critical insights into how PD progresses in real-life settings, including how patients respond to treatments over time and how care patterns differ across populations. The potential applications are wide-ranging:
- Identifying early disease markers: Through longitudinal analysis, the ability to gain detection of subtle changes and early symptoms that may precede a PD diagnosis—such as changes in gait, speech patterns, or handwriting—helps open the door for earlier interventions.
- Enhancing patient stratification: RWD allows for more precise segmentation of patient populations based on real-world phenotypes and disease trajectories, improving the design and targeting of clinical trials.
- Developing External Control Arms: With high-quality, regulatory-grade RWD, researchers construct external control arms that mirror clinical trial populations, potentially reducing the need for traditional placebo groups and making trials more ethical and appealing to patients.
- Evaluating long-term treatment effectiveness: By capturing outcomes across years, RWD supports post-market surveillance and helps assess how different therapies perform across diverse demographic groups in routine care settings.
This shift—from episodic, isolated trial snapshots to continuous, real-world insights— dramatically accelerates therapeutic discovery and enables more patient-centric research.
Artificial Intelligence: Unlocking Hidden Insights in Parkinson’s Disease
While the promise of RWD is vast, its sheer volume and variability pose challenges. This is where AI comes in. AI techniques, such as machine learning (ML) and natural language processing (NLP), can transform large-scale, complex datasets into actionable intelligence by detecting patterns and relationships that are otherwise difficult to identify.
PD is uniquely positioned to benefit from AI-powered insights. Much of the relevant clinical information–-such as descriptions of tremor severity, freezing episodes, or medication-related complications–lives in unstructured clinician notes rather than structured EHR fields. NLP extracts and standardizes these insights to create a fuller picture of a patient’s disease experience.
Key areas where AI can make a difference include:
- Early Diagnosis and Disease Prediction: AI models trained on multimodal data—such as clinician notes and imaging—can help identify early signs of PD before a formal diagnosis, potentially enabling interventions that delay progression.
- Personalized Treatment Planning: By analyzing large datasets, AI can uncover what treatments work best for which patients based on similar profiles, supporting more tailored and effective care.
- Clinical Trial Optimization: AI can help identify eligible participants faster and more precisely by sifting through unstructured data and matching patients to appropriate study criteria—speeding up recruitment and improving trial success rates.
The Path Forward: A Collaborative Approach to Innovation
The potential of RWD and AI in transforming PD research is immense—but unlocking these capabilities requires a coordinated effort. Collaboration across healthcare ecosystems is essential. Researchers, clinicians, life sciences companies, technology innovators, and regulators must work together to ensure data quality, safeguard patient privacy, and establish frameworks for validating and applying AI-driven insights responsibly.
Trust is also critical. Stakeholders need confidence that AI models are transparent, explainable, and built on representative, high-integrity data. This means adopting rigorous standards for data curation, bias mitigation, and continuous validation.
Through partnerships with leading specialty medical societies and deep expertise in structuring complex clinical data, it’s possible to build the evidence base for a future where PD care becomes more predictive, personalized, and proactive.
By embracing the power of RWD and AI, we can move beyond the limitations of traditional research and bring about meaningful breakthroughs for the millions affected by PD.
About Dr. Heather Moss
Dr. Moss is a medical advisor at Verana Health, as well as a Professor of Ophthalmology and of Neurology and Neurological Sciences at Stanford University. Dr. Moss pursued undergraduate studies in biomedical engineering at the University of Guelph, followed by doctoral studies in medical engineering at Harvard and MIT, seeking to improve human health through application of engineering principles. She has published over 100 articles in peer-reviewed journals, has authored numerous book chapters, and serves on the editorial board of four journals. Her clinical expertise includes diagnosis and treatment of optic nerve diseases, eye movement disorders, and neurological pathology affecting visual pathways. She is a fellow of the American Academy of Neurology and the North American Neuro-Ophthalmology Society and has been elected to leadership roles in both organizations.
